M
MITL Quant Alpha
Phoenix Equity Timing Strategy
PETS · Capital Deployment Architecture

A systematic framework for how capital is deployed, recycled, and compounded — not a signal service.

PETS is a vertically integrated, rules-based capital deployment engine built for institutionally tradable universes. Every layer — research conditioning, signal generation, portfolio construction, execution, and mandate overlays — is connected, automated, and publicly accessible for allocator due diligence.

~2,000 symbol universe 17 years of frozen rules Zero discretionary overrides Phoenix Capital Compounding™ Full audit trail Mandate-flexible
What this site is
An architecture-first presentation of PETS. Public pages describe structure, posture, and methodology. Performance validation is in the Institutional Snapshot — not surfaced here as marketing.
What is publicly accessible
Research conditioning across ~2,000 symbols. Backtest summary and institutional comparator context. Weekly regime posture. Live execution verification. All descriptive by design.
What is reserved for diligence
Backtest equity curves. Portfolio-level and single-symbol validation views. Proprietary thresholds, execution semantics, and reconstruction-relevant mechanics. Available under credentialed access.
Disclosure
Informational and educational only. Not investment advice. Not an offer or solicitation.
1 Integrated ecosystem
Architecture overview

Research conditions the engine. The engine drives execution. Execution feeds validation. The same architecture expresses itself under different mandates.

Every module is connected. None operates in isolation.

Most quantitative systems are point solutions — a signal feed, a screener, a backtester in isolation. PETS is a complete, connected ecosystem where each layer informs the next. The research conditioning layer determines which symbols enter the signal engine and at what priority. The signal engine governs weekly deployment posture. Portfolio construction compounds capital performance-responsively across winners and contracts it across underperformers without manual intervention. Execution verification confirms the strategy behaves as designed in live conditions. And the retirement overlay demonstrates the same engine operating under a capital-preservation mandate — without strategy modification. The depth is not in any one layer. It is in the coherence across all of them.

Layer 0 — Conditioning

Research Intelligence

~2,000 symbols

Multi-dimensional symbol conditioning: fundamentals, quality, growth, institutional ownership, PETS-specific edge, path quality. Output is strategy-conditioned compatibility — not a universal rank.

Fund → Quality → Growth → Owner → Edge → Priority → Selection
Descriptive · non-advisory Open
Layer 1 — Validation

Institutional Snapshot

Backtest · Comparator

Aggregate backtest summary. Institutional comparator context — 13F replication strategy, 50-fund scoreboard. Chain-of-custody execution audit. Designed for allocator diligence, not performance marketing.

17yr · Citadel Panel B · audit-trail orientation
Allocator-grade · neutral Open
Layer 2 — Mandate flexibility

Retirement Overlay

Capital-preservation mandate

The same PETS engine under a different mandate constraint. Weekly posture states: Accumulation, Participation, Caution, Defensive. Demonstrates mandate flexibility without strategy modification.

Dual-mode · 67 weeks live · 17yr validated
Non-advisory · plain language Open
Layer 3 — Diligence interface

Validation Engine

Credentialed ↗

Interactive backtester with full NAV tracking, Phoenix Capital Compounding™ portfolio construction, drawdown analysis, trade logs, and export artifacts. Designed for allocator due diligence review.

Live · paper trading · forward execution tracking
Login required Open
2 Architecture thesis
Signal architecture

Six layers from research conditioning to deployment priority — each empirically validated independently

Structural pillars
Fund · Quality · Growth · Owner Rank

Universal factors: fundamental quality, governance, growth consistency, institutional ownership behavior. Strategy-agnostic. These describe the business — not the trade.

PETS-specific layer
Edge Score

17-year historical fit between a symbol's behavior and the PETS regime/timing framework. Not a generic momentum factor — it measures extractability under this specific engine's mechanics. Out-of-sample persistence confirmed across all tested stress periods.

Capacity-aware composite
Priority Score

60% Owner Rank + 40% Edge Score. Empirically derived weighting validated across 7 independent stress-week periods from 2012–2025. Not tuned; tested and confirmed.

Deployment output
Selection Score

Path quality overlay applied to Priority Score. Final deployment rank. Top-10 candidates when PETS signals a deployment week. Capacity and liquidity constraints enforced throughout.

The distinction between universal pillars and PETS-specific edge is deliberate. Universal factors describe structure; edge describes extractability under this strategy's mechanics. Explore Research Intelligence →
Portfolio construction

Phoenix Capital Compounding™ — performance-responsive sizing without manual rebalancing

Most systematic strategies face an architectural tension: equal weighting ignores performance history, and volatility-targeting introduces parameter sensitivity. PETS resolves this through Phoenix Capital Compounding™ — each symbol maintains an independent capital account. Profits stay associated with the symbol that generated them. Losses contract that symbol's future allocation automatically.

The result is a performance-responsive sizing effect that operates without manual intervention, without explicit optimization, and without predictive forecasting. Capital naturally migrates toward symbols where PETS signals have historically been most effective. Weaker contributors scale down continuously — not at quarterly rebalance intervals.

This is a design choice made to reduce parameter sensitivity and allow durable signal persistence to express itself over multi-year horizons. The architecture remains intact across pooled funds, managed accounts, and internal strategy sleeves — external capital flows do not alter how the strategy constructs positions.

Phoenix Capital Compounding™ is a proprietary portfolio construction methodology. Reconstruction-relevant accounting mechanics are reserved for credentialed diligence review.
3 What this is — and what it is not
Positioning

A capital deployment engine — not a stock selection system, not a signal feed, not a screener

The distinction matters because capital deployment efficiency is the variable being optimized.
PETS is
Rules-based and deterministic
Every entry, exit, and sizing decision follows frozen rules. No version drift. No parameter updates since inception. Every decision reproducible and auditable.
Weekly cadence, multi-month holds
Institutional time horizon. Low turnover. ~30 trades per symbol across 17 years. Not latency-sensitive or capacity-constrained.
Strategy-conditioned compatibility
Candidates evaluated for fit with this specific engine's mechanics. A symbol's score is conditional on PETS' extraction process — not a universal rank.
Mandate-flexible architecture
The same engine operates under capital-growth and capital-preservation mandates. Overlay logic adapts posture without modifying the underlying strategy.
Empirically validated, component by component
7 major research studies. Each component tested out-of-sample. Components that failed testing were discarded and documented as such — not retained for completeness.
PETS is not
Not a signal service
No alerts, no recommendations, no "buy this now." The system generates internal deployment decisions — not signals distributed to subscribers.
Not a screener
A screener answers "what looks good universally?" PETS answers "what is structurally compatible with this engine, in this regime, under capacity discipline?"
Not black-box ML
Transparent signal logic. Interpretable rankings at every layer. No opaque neural networks. Allocator-defensible throughout.
Not market timing
Always invested (80–88% average exposure). Captures systematic opportunities within regimes — not macro calls about when to be in or out of the market.
Not dependent on cherry-picked names
Deliberately tested on "worst 10" symbols. Architecture does not depend on a favourable universe. Performance characteristics hold across sectors and symbol quality bands.
Public content stays descriptive. Thresholds, parameter values, and reconstruction-relevant execution mechanics are not disclosed publicly and are available under credentialed diligence.
4 Research depth — what was tested, what was rejected
Validation record

Seven major research studies. The findings that failed testing were discarded and documented — not quietly dropped.

Credibility comes from what was rejected, not just from what was retained.
1
Exit Overlay Study
Profit-protection exit overlays cut exposure to the payoff tail — reducing total PnL. Finding: exit logic remains at the portfolio level only. Position-level overlays rejected.
Overlay rejected
2
Entry Selection Study
Owner Rank is the only overlay that consistently improves results across all tested periods. Trade rank, chaos score, and structure score confirmed as diagnostic only — not selection criteria.
One factor confirmed
3
Ranking Mechanisms Study
Owner Score dominates all composite rankings across 7 independent periods. 0.60/0.40 Owner/Edge weighting empirically confirmed across stress weeks from 2012–2025. Validated, not tuned.
Architecture confirmed
4
Edge Score Credibility
Out-of-sample persistence confirmed across all tested stress periods including 2020, 2022, and recent drawdowns. Edge is not a backfit artifact — statistical persistence demonstrated.
OOS persistence confirmed
5
Rebound Pattern Study
Rebound patterns as mechanical trading rules do not increase expectancy. Retained as a contextual overlay only — not a trading signal. The distinction is material to the architecture.
Rejected as rule
6
Seasonality Overlay
Calendar effects provide tail-risk awareness without cutting convexity when used as a secondary, per-symbol filter only. Broad market timing application was tested and rejected.
Retained as filter
7
Internals Monitor
Post-entry institutional behavior (volume imbalance, absorption, AVWAP) confirms capital allocation quality. Position states guide ongoing sizing decisions — not entry criteria.
Post-entry confirmed
Full research documentation (15 PDFs) available for credentialed allocator review. Each study documents confirming and disconfirming evidence.
5 Governance and operational profile
Rules integrity

Frozen engine. No human intervention. No discretionary overrides. The same logic since inception.

PETS operates on a completely frozen rule set. The same logic that generated signals in 2008 generates signals today. No annual re-optimization. No parameter updates when performance dips. No PM judgment calls during drawdowns.

This is a governance design choice — not a limitation. Frozen rules enable genuine out-of-sample evaluation, eliminate key-person risk, and produce a verifiable 17-year audit trail where every decision can be traced back to the same deterministic logic. The backtester UI provides full decision audit exports for diligence review.

Deployment profile

Production-ready. Fully automated. 1–2 FTE operational overhead. Plug-and-play across implementation contexts.

Saturday signal generation. Sunday order staging and email distribution. Monday open execution via Alpaca. Forward execution tracking via GitHub. Weekly summaries to configured recipients. All automated.

Complete Python codebase. 1,000+ symbol pre-curated universe. Multi-vendor data validation across EODHD, Alpha Vantage, Yahoo Finance, and TradeStation. Configuration to paper-trading validation to live production in 1–2 weeks. No research team required — the research is complete. Phoenix Capital Compounding™ remains intact across pooled funds, managed accounts, and internal strategy sleeves.

Current status: live paper trading, Alpaca broker. Interactive Brokers integration in progress. VPS/Nginx migration in progress.
6 Allocator inquiry
Credentialed diligence · licensing · strategic partnership

If the architecture resonates, the next step is a deeper look — not a sales call.

PETS is not presented here to close a transaction. It is presented to demonstrate that serious, systematically-built, deeply validated capital deployment infrastructure exists and is ready for institutional review. If you are a CIO, allocator, or quantitative due-diligence team and the architecture warrants further evaluation, the full diligence package — 15 research PDFs, backtester access, live track record, and architecture documentation — is available under NDA.

Engagement structures
  • Strategic partnership aligned to deployment scale
  • Technology licensing — compiled core or full source
  • Revenue participation linked to strategy performance
  • Joint venture vehicle using your operational platform
What to include in your inquiry
Mandate type and capital range.
Preferred engagement structure or area of interest.
Liquidity and capacity requirements, if applicable.
Review scope — architecture overview, full diligence, or licensing discussion.
Contact
Rakesh Shah, Founder
MITL Quant Alpha  ·  quantalpha2025@gmail.com
Begin Allocator Review
Informational and educational only. Not investment, tax, or legal advice. Not an offer or solicitation to buy or sell any securities. Past performance does not guarantee future results. Backtest results are hypothetical and subject to inherent limitations. All commercial terms determined collaboratively following technical diligence and structural alignment.